1. Field of the Invention
The invention relates generally to systems and methods for calibrating and testing light-emitting apparatus for determining the range and/or velocity of remote objects and relates more particularly to calibrating a three dimensional light detection and ranging apparatus.
2. Related Art
There are a wide variety of applications for apparatus which determine the distance and/or the velocity of remote objects by means of detecting backscattered light. Laser radars (LIDARs) determine range in the atmosphere by measuring the transit time of a laser pulse from the transmitter/receiver to the target and dividing by twice the velocity of light in the atmospheric medium. Range resolution in such devices is related to the accuracy of this transit time measurement. In the atmosphere, ranges are typically measured in kilometers, where range resolution can be as small as 30 cm. A 3D target image can be obtained with a laser radar by rastering the laser beam across the target and measuring the transit time, pulse by pulse, where each pulse corresponds to a point on the target. The distance between points on the target determines the spatial resolution of the rastered image and defines the picture element (pixel) size; the number of pixels at the target determines the pixel-array size; the range resolution determines resolution in the third target dimension. Rastering is a slow process, particularly for large pixel-array sizes, and it requires cumbersome mechanical scanners and complex pixel-registration computer processing.
When high-speed imaging is required, it is desirable to obtain the entire image with one laser pulse. However, because of weight, cost and complexity problems, it is undesirable to obtain the entire image with independent, parallel, laser radar receivers, where each pixel uses a separate laser radar receiver system; multiplexing with one laser radar receiver should be used to make a large pixel-array imaging system practical. Currently lightweight, multiplexing laser radar receivers (Flash LIDAR) exist which can image an entire target with a single laser pulse. It accomplishes this feat by having the returning pulse stop a clock at each pixel. This clock can be a voltage ramp; that is a voltage which is decreasing in time. The voltage at which the ramp is switched off determines the time at which the laser pulse returned. The voltage ramp begins when the laser pulse is transmitted. Because stopping a clock or switching off a voltage ramp is usually dependent upon the returning-pulse amplitude for typical laser pulse widths (due to the clock-stopping circuitry) and the returning-pulse amplitude is dependent, pixel by pixel, on the reflectivity of the target, the range must be corrected for amplitude to obtain high range resolution. By means of switching off a ramping voltage, individually for each object pixel, the time of arrival of the reflected pulse is recorded. This time is related to the third object dimension. The device consists of the pulsed light source, optics for collecting the reflected light, a sensor for detecting the light, drive and output electronics for timing and signal conditioning of data generated by the sensors and a computer for processing the sensor data and converting it to a three dimensional image. The sensor collects and processes the light data in a unique manner, using a hybrid. The hybrid is actually a two dimensional array of collectors or detectors combined in very close proximity with their own processing electronics. In general, the hybrid is composed of two integrated circuit chips mated together. The two dimensional array defines two dimensions of the image. The processing electronics individually and independently switch off, at high-speed, a time varying voltage when the light pulse, reflected from the object, arrives at the sensor. The final voltages are stored and can be mathematically transformed to the third object dimension. The sensor also records peak light pulse information which can be used to obtain higher resolution in the third dimension.
Precise calculations of range, velocity, and other factors determined by the Flash LIDAR apparatus require that the LIDAR apparatus be precisely calibrated throughout the useful life of the apparatus. Conventionally, the calibration process is labor intensive. A lengthy testing range is prepared by providing a fine optical alignment between the Flash LIDAR apparatus and a distant “hard target” which consists of a three dimensional object. The return light signal scattered back from the hard target is received by the apparatus and is compared to expected signal characteristics of the target, the range geometry, and the atmospheric conditions. If the actual return light signal is different than the expected light signal, the apparatus is adjusted accordingly.
This conventional calibration approach often requires obtaining a right of access and a permit to transmit the laser beam over the range property or construction of a large and costly facility. Moreover, at least two persons are required during the initial alignment of the hard target, since the hard target may be one kilometer from the LIDAR apparatus. Another concern is that inclement weather will adversely affect the calibration process, since fog, rain and snow will introduce backscattering and severely attenuate the beam intensity over the open range. Furthermore, unless the output beam is eye-safe, safety measures must be used to assure that no person in the vicinity of the testing range receives excessive eye exposure to the laser beam energy. The conventional calibration process is applied to ocean-going LIDAR imaging in U.S. Pat. No. 5,311,272 to Daniels et al. The patent describes deploying air-dropped buoys for use as calibrated targets for a system which images submerged objects. The calibrated optical buoys may be deployed by a helicopter.
U.S. Pat. No. 5,264,905 to Cavanagh et al. describes automated test equipment that overcomes many of the concerns with the conventional calibration process. A portion of an output beam from a laser rangefinder is directed to an inlet port of an integrating sphere which disperses the beam for exit through two outlet ports. The first outlet port is connected to a radiometer for measuring the dispersed energy. This measurement is used to determine the energy output of the laser rangefinder. The second outlet port communicates with an avalanche photo diode (APD) that detects the pulse envelope of the laser beam. The pulse envelope is connected in parallel to two separate measuring circuits for determining the pulse width and the pulse interval. A third connection from the APD is made to a pulse delay generator which generates a delayed trigger signal. The delayed trigger signal is input to a laser diode that generates a laser pulse to simulate the return of backscattered energy to the laser rangefinder. The laser pulse that is generated by the laser diode is directed to the integrating sphere for return to the laser rangefinder. The duration of the time interval between the sending of the signal from the laser rangefinder to the reception of the artificially generated return signal is calculated and used as the basis for calibrating the LIDAR apparatus. The pulse delay generator is programmable, so that the time interval can be adjusted to provide a more thorough calibration.
The automated test equipment of Cavanagh et al. overcomes many of the concerns associated with the conventional calibration process. The automated test equipment may be used indoors and requires less space. The laser diode that is triggered to simulate a return signal has sufficient intensity to be reliably detected by the laser rangefinder. However, in calibration of a LIDAR velocity measurement system, as previously noted, the simulated signal must be coherent with the frequency shifted by only a small fraction from the center frequency of the LIDAR transmitter. A triggered laser diode as described by Cavanagh et al. does not possess the required coherence, nor a center frequency that is close enough to the transmitter center frequency to accomplish a velocity calibration. In addition, in some calibration operations, a strong signal return such as that provided by a triggered laser diode may not be desirable. For example, if the LIDAR apparatus is to be used to measure atmospheric parameters, such as wind velocity or wind velocity distribution, accurate calibration may not be achieved when the return signal from a hard target or the laser diode of Cavanagh et al. is many orders of magnitude higher than the return signal that will be received in the actual application. This is because, in the practical application, the return light signal is the low level backscattered light from small aerosol particles entrained in the air, at some distance from the receiver. The receiver optics and electronics are optimized to detect these small levels, and could be saturated or damaged by very high signal levels. In addition, because the actual signal levels are so low, it is customary to employ signal pulse accumulation over many hundreds of successive output pulses, followed by sophisticated signal processing circuits and software to extract the velocity information. The high level signals provided by hard targets or triggered laser diodes do not allow the “end-to-end” testing of the receiver system through the signal processing electronics and software.
Another concern in the Cavanagh et al. approach is that the calibration of the laser rangefinder is dependent upon the proper calibration of the automated test equipment. For example, if the pulse delay generator is improperly calibrated, the laser rangefinder will be inaccurately calibrated. There must also be a precise time-compensation for the delays caused by the electrical operations that occur for artificially generating the return light signal. The return light signal is generated by the equipment only after operation of the APD, the pulse delay generator, and the laser diode.
U.S. Pat. No. 5,825,464 to Feichtner solves most of the previous calibration problems by providing a physical fiberoptic delay line with an integrating sphere input to the LIDAR receiver that uses the actual laser output from the LIDAR transceiver. This device only generates one dimensional range data.
None of the previous systems are capable of providing a true series of two dimensional range gate images to simulate a three dimensional LIDAR object.
What is needed is a system and method for calibrating and testing a light-emitting apparatus for detecting remote three dimensional objects by means of backscattered light, with the system and method being applicable to achieving thorough calibration over a wide range of return light signal strengths as well as simulating static and dynamic target conditions.